The Mutual Fund Show: Follow These ‘Good Rules’ To Earn Higher Returns
Here’s how rule-based investing can help mutual fund investors earn higher returns.
Patience is the key to success in investing as volatility in stock market can stir costly emotions—fear, greed or pride. A set of proven rules can guide investors through the ups and downs in the market.
Rule-based investing—a unique, yet effective strategy to invest in mutual funds where investors follow a set of time-tested, good investing principles to garner high returns.
“This kind of rule-based approach aims to deliver alpha over the benchmark in a very cost-efficient way, following rules which have been tested over time and geographies,” said Aparna Karnik, vice president at DSP Investment Managers.
Agreed Kalpen Parekh, president at the fund manager. “Good rules are those that win the long run,” he said in BloombergQuint’s weekly series The Mutual Fund Show. “Good rules are those that win more often. Just like the best batsman doesn’t score a century every time, good rules can sometimes underperform. But in the long run they will always help increase returns.”
Parekh also said a model designed and implemented at DSP has provided returns across market cycles.
The model includes these steps:
- Elimination: Remove weak companies based on set elimination criteria out of the universe of BSE200.
- Selection: Pick companies after applying growth, quality and value screeners.
- Weights: Give weights to respect the benchmark weights of the companies.
- Good Rules: Apply good rules to the remaining companies and stick to them without getting human emotions and biases in the way.
“Biases act as deterrents to alpha generation,” Parekh said.
Watch Parekh and Karnik decode and simplify this technique of investing in mutual funds:
Here are the edited excerpts of the interview:
How is it that investors would benefit from rule-based investing? Is this principle applicable for mutual fund investing as well?
Aparna Karnik: This kind of rule-based approach basically aims to deliver alpha over benchmark in a very cost-efficient manner, following rules based on evidences tested over time and geographies, based on research and without any kind of human emotion. So, there are various points where we would like to differentiate these kinds of funds from traditional active equity fund.
One, these rules are followed without any discretion or bias. Second, this fund is good for investors. We believe it is a complementary strategy. There’s a role for active, passive and moral-driven strategies to be part of an investor’s portfolio. These strategies should be very cost-efficient to achieve certain allocations or for investors to express certain investment philosophies such as those who would like to have quality-oriented or a growth-oriented portfolio. There is a combination of quality, growth and value, which is very systematically built in their portfolios.
Is there a broader thought that if there is rule-based investing and therefore taking the human intervention out, the results because of cost efficiency would be much better over a longer period of time? It may not happen every year, but this rule-based approach could hold investors in good state.
Kalpen Parekh: Absolutely. That is the clear reason as why we tested the efficiency and consistency of this rules and their ability generally to beat the benchmarks over a long period of time across three different market phases.Good rules are those that win in the long run and win more often. In cricket too, the best batsmen will not score a century every match. Good rules temporarily can underperform.
We worked around a year on the design of these rules. The starting intent was if the benchmark generates X percentage average returns over a full cycle, there are companies which generates more than X and companies which will generate less than X. Then we looked back that what are the characteristics of the losers. We said lets model for that and eliminate such companies. So, the first stage of rule-based approach is elimination. From the BSE 200 companies, we first have conditions which will eliminate companies.
There are three conditions. First, companies with high leverage.Because eventually in the long run and in a country like India where cost of capital is high, companies which needs more and more debt to grow their business somewhere meet accidents and do not create value, earn less than benchmark returns. Then we tested for this over the last 16-17 years since where the data is available. Also, these rules have been tested in global markets where the history of data is larger.
Second, companies where stock price beta is very high, which means the stock price goes up more than the market in a rising market and fall more than the market in a falling market.
Third, companies where capital allocation, management,governance issues are not friendly to shareholders. When you eliminate these then that itself gives you 2 percent alpha over a 14-year time horizon. That’s the first step.
Then we looked at the characteristics of companies that have created alpha, not just in one-two years but also over long period of time and across market cycles. These are generally companies which have good management, have return on capital and return on equity much higher than the cost of capital, that don’t need excessive debt to grow their business and have stability in their earnings.
We looked at ROEs, stability of earnings and future earnings growth expectations, and two value factors—price-to-free cash flow and dividend yield. So, you choose quality companies which have growth dimensions. You have growth and then you choose companies with reasonable valuations.
We gave weights and scores to each company on these five factors. These five factors come in three buckets of quality, growth and valuation.So, out of BSE 200, we eliminated 100 companies. Then we ran forensic and accounting screeners to test for poor accounting, governance issues, mismatch in data points in balance sheet or earnings report. That eliminated a lot of companies. After that, around 75 companies were left that were scored on growth,quality and value. And companies which captured the top 50 percentile went into the portfolio. So, the second step is selection.
The third step is what weights do you give. This puts together portfolios of around 45 companies. Then they were given weights which is important because you want to ensure that broadly they respect the benchmark weight of these companies and they respect the liquidity of these companies. They may be good companies, but liquidity is poor in the market and then concentration risk is important.
How do any of the funds which follow this practice differentiate themselves from any other well-run fund?
Kalpen Parekh: So far, we said about the art part of the selection style of fund manager and elimination style. Then we are bringing in the science part that sometimes personal human emotion and bias come in the way.
For example, for every fund manager there is daily tracking of NAVs, quarterly tracking of ranks and scores and there is peer group performance pressure. So, there are human biases which may come in. There could be short term trends which will be different than the investment principles you generally believe in. Fund managers try to adjust their portfolios to near term trends but sometimes that works and sometimes doesn’t. This set of rules-based funds say that you don’t want to create these set of deviations and you have to stick to these rules forever, so what if short-term trends trigger. Typically, there are years in which there is a very high beta rally. Lot of companies with leverage do very well. Companies which in the long run have not created value or alpha go up 1-2 times in the next 1.5 years. There are rallies in commodity cycle and commodity stocks run up. Here you need to time them. In rule-based fund, we say that some nature of businesses within the long run are not value creative and they will never enter the portfolio. And hence I don’t need to time them in or out.
So, you are eliminating personal and human biases. Biases sometimes acts as a deterrence to alpha generation. In times like today where benchmarks are becoming more efficient, how do you design a passive-plus strategy. So, this is passive-plus strategy.
In lot of years, this model would have surpassed the returns. When you look at the kind of returns which have come in through this model and the characteristics to choose the companies which will enable you to reach this, are they a defined set or are they left in some part to a broad range within which a fund manager can operate?
Aparna Karnik: In this case, it is specifically defined. If you look at it in a broader way, there are various macro conditions under which different factors or styles do well. So, the macro conditions and sentiments also determine which type of style and fund is being rewarded in the market. There are times of risk aversion where quality companies tend to outperform. There are times of more overheating conditions or growing economy where growth and value companies tend to do better.
We are saying that we have designed a model which captures all three dimensions and by doing so we are confident that based on whatever research and data we tested over the years, we will follow it very rigorously without any deviation, without any fund manager’s discretion. We believe for an investor who will stay invested for a medium term, the probability of generating alpha over benchmark is 92 percent. If you are ready to stay for five years, then the probability of beating the benchmark is 100 percent. That’s what we recommend and expect.
Is it essentially the lower expenses plus focused rule-based approach which either direction having surpassed would cut the position or do it?
Kalpen Parekh: Based on 15-years data, we tested in three phases where 2004-2008 was a massive bull run when the benchmark went up 7-10 times. Then 2008-09 was a sharp correction and then a period of no returns for five years. Till 2013, markets went nowhere. Last five years have been an up and down move with a gradual 10 percent CAGR return. In all these three pockets, does the model work?
What we are going through is very simple principles of investments—what are bad companies, what parameters signify bad companies and eliminate them. Also, what are good companies, put them in rules and don’t deviate them from them.
The hero of this is the design of the model. The human intervention is in the model design. Once the model has been designed, there is no human intervention. Then we strictly follow it. In the last couple of days, this portfolio could have underperformed as there has been a sharp beta rally as well. So, there are periods when such beta rallies happen, and this model will underperform.
Can you explain some of the rules?
Kalpen Parekh: We look for three broad factors— quality, growth and value. This is the first attempt in India of multifactor investing. Globally, factor investing is very popular. In India, NSE and BSE have come up with factor investing but they are largely single factors. So, there are indices which are quality factor, low volatility factor and value factor.
The two rules for screening good companies are those with highest ROEs and with highest consistency of earnings growth—they come under quality. Under growth, the rule expects future earnings growth. Under valuations, the rules are price-to-free cash flow and dividend yield. These are the factors which we use for elimination. We have a 22-factor forensic screening model. No company will be completely free and will have one-two red flags. Companies which are weak on data points get screened out.
When can such funds underperform?
Aparna Karnik: The way we have designed this fund is using a very strict elimination criterion. In a sense what are we eliminating is highly leverage companies, highly volatile companies. So, under market conditions where highly leverage companies and highly volatile companies are doing better, which is when there is a junk rally or kind of euphoria driven rally, in such a circumstance the fund may temporarily underperform.
Any active fund or any smart beta strategies whether it is single factor or multi-factor will have periods of outperformance and underperformance. You will never have a fund which will never underperform. However, we believe that such irrationality may not persist for very long and eventually fundamentals will catch up and hence over a slightly longer-term horizon, market will reward to rationality and the funds will deliver better returns.
We are specifically not considering any technical momentum-based signals that way we have designed this portfolio. That’s because we wanted it to reflect fundamental principles and not be based on technical sort of price-related aspects. So, if there’s a very sharp momentum rally in any stocks which is not backed by any fundamentals, the model may not be able to pick it up, it may pick it only when it is broadened out eventually by numbers. No fund can pick up each and every source of alpha. So, in this fund we are trying to create a portfolio of good, healthy, profitable, well-run, dividend paying companies which will compound wealth over time and deliver a reasonable alpha to the investors.
If one applies these eliminating criteria, the pool which will be left will be so small and would be so overvalued, that you will get such companies at favorable price only when the markets have cracked significantly and there is fear on the street. Otherwise you will have to buy them expensive.
Kalpen Parekh: Today the portfolio is more expensive than its benchmark. And for the last three-four years the market has rewarded quality a lot more. The recent valuation of portfolio is at a premium to the benchmark. If you look back 20 years, quality companies had always been at a premium to the benchmark. The premium has only expanded right now which could also be the source of near-term underperformance. But I can’t predict that. Having said that, companies which have durable long-lasting performance, at the end of the day are same groups of the companies. Every two-three years you do see some events where there is a price correction in the market. So, every two to three years you have 10-15 percent usual correction in the market. Every five years there is 20-25 percent drawdown in the market. Every once in 10 years there is 25-30 percent correction in the market. Even good companies come down and we advise investors that if you intend long term wealth creation for retirement, you should be happy when prices come down. What you are sure of here is that these are the companies which hopefully will not die because you have tried to eliminate most of the factors which can be the source of accidents.
You have an NFO which is opening based on this model.
Kalpen Parekh: It has opened on May 20. It’s called DSP Quant Fund.
Any other example of such thing or is this the first coming to India?
Kalpen Parekh: There are rule-based asset allocation funds in the country but once the asset allocation is done,the selection of stocks is still human where the fund manager will choose active stocks and decide exposure between different sectors. A completely hardcoded, multi-factored, rule-based fund, this is the first one.
So, therefore by nature the expenses to manage such funds are lower?
Kalpen Parekh: Should be lower and hence, are lower.
In our many conversations with people, for instance Saurabh Mukherjea had said in lot of cases the expenses amount over a period. They were trying to say this between regular and direct investment, how 20-35-year period can make a big difference.
Kalpen Parekh: Expenses matter. The point in India is one thing which gets missed out in this whole debate is that less than 3 percent of population is invested and for them more than expenses getting into the asset classes is more important.
And since you mentioned Sourav, he has been running these screener-based portfolios and has these good and clean portfolios. They are on similar lines of elimination and selection but after that there is still a manual intervention. But we are saying that once the model is designed then there’s no manual intervention. In that context, this is the first of its own kind.
So, both elimination of stocks from the portfolio.
Kalpen Parekh: From the benchmark.
Once the portfolio is screened then at some point of time may be the company’s parameter change.
Kalpen Parekh: They get eliminated.
All of it will be automated in that sense. So, human intervention is minimal or zero?
Kalpen Parekh: Its zero.
Aparna Karnik: The human intervention is only running the model, executing and continuously accessing health and robustness of the model. The running and execution will be based on the model without discretion.
Yesterday one of the NBFCs said we are not allowing premature withdrawals. A lot of people are trying to figure out whether this will have a big impact on banks which might have these allocations to these NBFC papers and mutual funds as well. Is this a big worry or do you think not quite?
Kalpen Parekh: We would be anxious because as mutual fund industry there will be exposure to multiple NBFCs. This is a first incidence. What we hear is this the measure to probably ringfence liquidity and I think the company is also in talks toward bringing strategic partnership and there are some large payments in June. So, this in a way protects liquidity for those payments which are more in the nature of institutional bonds. We will have to wait and watch how it plays out.
Several brokerages have written reports on how this wave 2 might slightly be different from wave 1. Any thoughts?
Kalpen Parekh: It is always in hindsight that this was wave 1 or 2 or 3 but right now we are living through the period of tight liquidity, concerns on credits in various instruments and tight liquidity and high leverage is a matter of concern. These are cycles we see once in every five to six years, but I can only tell investors that this is the part of the investing journey. You always have one-two bad years in a five-year cycle. We have seen this in 2008 and 2013. So typically, in fixed income there is a five-year phase where liquidity tightens, rates tighten—either it’s an interested spike or it’s a credit event. So, the industry is also maturing and navigating through it. I am confident that as an industry collectively we will pass through this and get into the next cycle.